INDEX

Explanations

elements related to managing requests and responses in a programming context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

verification tokens

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SharedDtor

-0.59

 متعلقه

-0.55

Ƚ

-0.54

 Chwiliwch

-0.52

󠁣

-0.47

 AssemblyCulture

-0.47

Personensuche

-0.45

Ecotoxicity

-0.44

 Houſe

-0.43

ématiques

-0.42

POSITIVE LOGITS

tonode

0.49

 فريبيس

0.40

 ditch

0.37

bad

0.36

$_['

0.35

무

0.35

 scramble

0.35

 colorWith

0.35

تقاوى

0.35

 transpired

0.34

Activations Density 0.018%

elements related to managing requests and responses in a programming context

verification tokens

No Comments

No Known Activations